Angiogenesis has been mostly studied by examining stimulation (or inhibition) of endothelial cells (ECs) by extracellular ligands, with much less attention being paid to communications between ECs and their surrounding tissue environment. In this project we will focus on how the oxygen sensing mechanism coordinates astrocytic and vascular communication during retinal vascular morphogenesis and explore its therapeutic potentials in protecting retinal microvascular integrity in two models including oxygen induced retinopathy (OIR) and diabetic retinopathy. In preliminary studies we obtained concrete evidence that nascent retinal blood vessels drive retinal astrocytic progenitors (APCs) to differentiate into mature astrocytes (mASCs) and form an astrocytic network. Specifically, we hypothesize that leukemia inhibitory factor (LIF) from retinal vascular ECs and oxygen from the circulation may act on APCs to upregulate the expression and activity of prolyl hydroxylase domain protein 2 (PHD2), which in turn catalyzes prolyl hydroxylation of HIF-a proteins, tagging them for polyubiquitination and proteasomal degradation. Furthermore, we propose that HIF-2a and Tlx (a transcription factor in the orphan nuclear receptor family) may form a positive feedback loop wherein Tlx suppresses polyubiquitination and degradation of hydroxylated HIF-2a whereas HIF-2a promotes the transcription of the Tlx gene. Thus, downregulation of HIF-2a protein abundance by PHD2 might trigger a downward spiral of both HIF-2a and Tlx levels, leading to astrocyte maturation from their progenitors. On the other hand, our studies indicate that loss of PHD2 may be targeted to help preserve retinal vascular integrity in OIR or diabetes models, likely by stabilizing HIF-2a. We will investigate these issues in three specific aims. Aim 1. Determine if LIF regulates astrocyte and vascular development through oxygen sensing mechanisms. We will evaluate in vivo if LIF upregulates Phd2 expression, promotes the differentiation of APCs (proangiogenic) to mASCs (nonangiogenic), thus suppressing retinal angiogenesis. Aim 2. Investigate if a positive feedback loop between HIF-2a and Tlx promotes APC status and retinal angiogenesis. Aim 3. Explore therapeutic potentials of the oxygen sensing pathway in mouse OIR and diabetic retinopathy models. We will determine if mASC specific PHD2 deficiency, global or EC specific LIF deficiency, or LIF antagonist protects retinal microvessels from hyperoxia or diabetes-induced injuries. In summary, these studies are designed to provide mechanistic insights into vascular and astrocyte communication and reveal novel therapeutic opportunities to treat retinopathy of the prematurity and diabetic retinopathy.